>> diff drivers/firmware/imx/ele_common.c
...
>> + err = -EPERM;
>> + goto exit;
>> + }
>> + }
>> + set_se_rcv_msg_timeout(SE_RCV_MSG_LONG_TIMEOUT);
> you set default TIMEMOUT at suspend, but set LONG TIMEOUT here.
> What's happen, if one instance already resumed, and set
SE_RCV_MSG_LONG_TIMEOUT
> the other instance still in suspend, which will wrong use
SE_RCV_MSG_LONG_TIMEOUT.
You are correct. Since the interface is single here. It is not a problem.
For multiple interfaces, It is required to be put in priv structure.
Will update this in v15.
>> +
>> + err = ele_msg_send_rcv(dev_ctx,
>> + tx_msg,
>> + cmd_snd_rcv_rsp_info.tx_buf_sz,
>> + rx_msg,
>> a/drivers/firmware/imx/se_ctrl.c b/drivers/firmware/imx/se_ctrl.c
>> index 144a727bbfe0..9dbe866b7b13 100644
>> --- a/drivers/firmware/imx/se_ctrl.c
>> +char *get_se_if_name(u8 se_if_id)
>> +{
>> + switch (se_if_id) {
>> + case SE_TYPE_ID_DBG: return SE_TYPE_STR_DBG;
>> + case SE_TYPE_ID_HSM: return SE_TYPE_STR_HSM;
> If only use once, needn't define macro, just
> return "dbg".
It will be used multiple times in coming patches.
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
